chat接口改为支持stream

chatgpt-next-web/chat_proxy.py

@@ -0,0 +1,55 @@
+import websocket
+import json
+import os
+
+MY_OPENAI_API_KEY = os.environ.get('MY_OPENAI_API_KEY')
+
+
+def on_message(ws, message):
+    response = json.loads(message)
+    # 处理来自 ChatGPT 的响应
+    print(response['choices'][0]['text'])
+
+    # 根据您的条件判断是否需要关闭连接
+    if '完成对话的条件':
+        ws.close()
+
+
+def on_error(ws, error):
+    print(error)
+
+
+def on_close(ws):
+    print("连接已关闭")
+
+
+def on_open(ws):
+    # 发送初始消息
+    ws.send(json.dumps({
+        'type': 'input',
+        'message': '你好，ChatGPT！'
+    }))
+
+    # 发送后续消息（用于完成）
+    ws.send(json.dumps({
+        'type': 'input',
+        'message': '技术问题是...'
+    }))
+
+
+if __name__ == '__main__':
+    # 创建 WebSocket 连接和其他初始化代码
+    ws = websocket.WebSocketApp(
+        'wss://api.openai.com/v1/streaming_chat_completion',
+        on_message=on_message,
+        on_error=on_error,
+        on_close=on_close,
+        header={
+            'Authorization': 'Bearer YOUR_API_KEY'
+        }
+    )
+
+    ws.on_open = on_open
+
+    # 运行 WebSocket 连接
+    ws.run_forever()

redis/test_user_redis.py → chatgpt-next-web/service.py

@@ -7,15 +7,15 @@ import jwt
 import datetime
 import requests
 import os
-from flask import Flask, request, jsonify, Request
+from flask import Flask, request, jsonify, Request, Response
 from redis import Redis
 
 SECERT_KEY = "8U2LL1"
 MY_OPENAI_API_KEY = os.environ.get('MY_OPENAI_API_KEY')
 
 app = Flask(__name__)
-redis = Redis(host='192.168.3.229', port=6379, password='lizhen-redis')
-# redis = Redis(host='10.254.13.87', port=6379)
+# redis = Redis(host='192.168.3.229', port=6379, password='lizhen-redis')
+redis = Redis(host='10.254.13.87', port=6379)
 # redis = Redis(host='localhost', port=6379)
 
 
@@ -219,7 +219,7 @@ def validate():
     return jsonify({'code': 0, 'message': 'Chat limit not exceeded'})
 
 
-@app.route('/chat/completions', methods=['POST'])
+@app.route('/v1/chat/completions', methods=['POST'])
 def proxy_chat_completions():
     token = parse_token(request)
     model = request.json.get('model')
@@ -248,23 +248,40 @@ def proxy_chat_completions():
     # 获取请求数据
     data = request.get_json()
 
+    stream = request.json.get('stream')
+
     # 设置请求头部信息
     headers = {
         'Authorization': f'Bearer {MY_OPENAI_API_KEY}',
         'Content-Type': 'application/json'
     }
-    # 将请求转发到 OpenAI API
-    response = requests.post(
-        'https://api.openai.com/v1/chat/completions', json=data, headers=headers)
 
-    # 获取 OpenAI API 的响应数据
-    result = response.json()
+    if stream:
+        # 将请求转发到 OpenAI API
+        response = requests.post(
+            'https://api.openai.com/v1/chat/completions', json=data, headers=headers, stream=True, verify=False)
 
-    user_package_key = f'user:{user_id}:package'
-    redis.hincrby(user_package_key, 'basic_chat_limit', -1)
+        # 获取 OpenAI API 的响应数据
+        result = response.iter_content(chunk_size=8192)
+
+        user_package_key = f'user:{user_id}:package'
+        redis.hincrby(user_package_key, 'basic_chat_limit', -1)
+
+        # 返回 OpenAI API 的响应给客户端
+        return Response(result, content_type=response.headers['content-type'])
+    else:
+        # 将请求转发到 OpenAI API
+        response = requests.post(
+            'https://api.openai.com/v1/chat/completions', json=data, headers=headers)
+
+        # 获取 OpenAI API 的响应数据
+        result = response.json()
+
+        user_package_key = f'user:{user_id}:package'
+        redis.hincrby(user_package_key, 'basic_chat_limit', -1)
 
-    # 返回 OpenAI API 的响应给客户端
-    return result, response.status_code
+        # 返回 OpenAI API 的响应给客户端
+        return result, response.status_code
 
 
 def parse_token(request: Request):

openai-example/ stream_chat_completion.py

@@ -0,0 +1,17 @@
+import openai  # for OpenAI API calls
+
+# Example of an OpenAI ChatCompletion request with stream=True
+# https://platform.openai.com/docs/guides/chat
+
+# a ChatCompletion request
+response = openai.ChatCompletion.create(
+    model='gpt-3.5-turbo',
+    messages=[
+        {'role': 'user', 'content': "写一个50字左右的小故事"}
+    ],
+    temperature=0,
+    stream=True  # this time, we set stream=True
+)
+
+for chunk in response:
+    print(chunk)

openai-example/typical_chat_completion.py

@@ -0,0 +1,24 @@
+import openai  # for OpenAI API calls
+import time  # for measuring time duration of API calls
+
+# Example of an OpenAI ChatCompletion request
+# https://platform.openai.com/docs/guides/chat
+
+# record the time before the request is sent
+start_time = time.time()
+
+# send a ChatCompletion request to count to 100
+response = openai.ChatCompletion.create(
+    model='gpt-3.5-turbo',
+    messages=[
+        {'role': 'user', 'content': 'Count to 100, with a comma between each number and no newlines. E.g., 1, 2, 3, ...'}
+    ],
+    temperature=0,
+)
+
+# calculate the time it took to receive the response
+response_time = time.time() - start_time
+
+# print the time delay and text received
+print(f"Full response received {response_time:.2f} seconds after request")
+print(f"Full response received:\n{response}")